0.6.16
版本发布时间: 2024-08-05 23:49:41
langgenius/dify最新发布版本:0.8.2(2024-09-13 13:45:51)
✨ What’s New in v0.6.16? ✨
Hey everyone, we’re back with another update, and this one’s packed with some exciting new features, model support, and enhancements. Let’s dive right in!
🚀 Phasing Out N-to-1: Upgrading Multi-path Knowledge Retrieval
[!IMPORTANT] We're phasing out the
N-to-1
retrieval strategy on September 1, 2024, and introducing a more flexible multi-path retrieval strategy. We recommend switching to this new approach to boost your application's retrieval efficiency.
Learn More: https://dify.ai/blog/dify-ai-blog-n-to-1-knowledge-retrieval-legacy
🧠 Models
Llama3.1 Support
- OpenRouter Provider: Added support for Llama3.1 series models by @sinomoe in #6595.
- GroqCloud: Llama3.1 series models are now supported by @takatost in #6596.
- Bedrock: Llama 3.1 support added by @longzhihun in #6645.
- NVIDIA: Llama3.1 model support by @sulgweb in #6844.
GPT-4o Mini Support
- Azure GPT-4o Mini: Added support for Azure GPT-4o Mini by @charli117 in #6862 and @leslie2046 in #6866.
Other Models
- DeepSeek-Coder-v2: Added in SiliconFlow by @senseb in #6149.
- Novita Model Provider: New model provider added by @jasonhp in #6717.
- Solar Model Provider: Added new provider Solar by @JuHyung-Son in #6884.
- Hunyuan-Embedding: New model support by @maybemaynot in #6657.
- Xinference TTS: Added support for text-to-speech by @Weaxs in #6746.
- Tool-Call for Hunyuan: Added support for tool-call by @maybemaynot in #6656.
🛠️ Tools
- AWS Built-in Tools: Added by @ybalbert001 in #6721.
- Serper for Google Search: New tool added for Google search by @hwzhuhao in #6790.
⚙️ Enhancements
- Python 3.12 Support: Added Python 3.12 support for API Service by @bowenliang123 in #6771.
- Milvus 2.4.x Support: Initial support for Milvus 2.4.x by @bowenliang123 in #6084.
- Docker-Compose Certbot Configurations: Added with backward compatibility by @k-brahma in #6702.
- Celery Auto-Scale: Added support for Celery auto-scaling by @ZuzooVn in #6249.
- User Session ID Search: Enhanced App Logs search capabilities by @ZhouhaoJiang in #6638.
🧹 Cleanups
- Remove TTS (Blocking Call): Removed blocking call for TTS by @ic-xu in #6869.
For additional enhancements and updates, refer to the What's Changed section below.
Upgrade Guide
Docker compose deployments
[!WARNING] The
docker-compose.yaml
has been refactored. If you've made any changes to the file, make sure to check out the "Upgrade to new docker compose deployment" section above for usage and migration tips.
-
Back up your customized docker-compose YAML file (optional)
cd docker cp docker-compose.yaml docker-compose.yaml.$(date +%s).bak
-
Get the latest code from the main branch
git checkout main git pull origin main
-
Stop the service,Command, please execute in the docker directory
docker compose down
-
Back up data
tar -cvf volumes-$(date +%s).tgz volumes
-
Upgrade services
docker compose up -d
Source Code deployments
-
Stop API server, Worker and Web frontend Server.
-
Get the latest code from the main branch:
git checkout main git pull origin main
-
Update Python dependencies:
cd api poetry install
-
Then, let's run the migration script:
poetry shell flask db upgrade
-
Finally, run API server, Worker and Web frontend Server again.
What's Changed
- feat: support llama3.1 series models for openrouter provider by @sinomoe in https://github.com/langgenius/dify/pull/6595
- feat: add GroqCloud llama3.1 series models support by @takatost in https://github.com/langgenius/dify/pull/6596
- fix: downgraded sentry-sdk to 1.44.1 due to claude LLM token returning 0 by @takatost in https://github.com/langgenius/dify/pull/6597
- Feat/delete single dataset retrival by @JohnJyong in https://github.com/langgenius/dify/pull/6570
- feat: n to 1 retrieval legacy by @zxhlyh in https://github.com/langgenius/dify/pull/6554
- Fix: reset button in embedded chatbot by @JzoNgKVO in https://github.com/langgenius/dify/pull/6611
- Fix: DSL backup by @JzoNgKVO in https://github.com/langgenius/dify/pull/6616
- Fix/6615 40 varchar limit on model name by @crazywoola in https://github.com/langgenius/dify/pull/6623
- fix: qwen fc error by @Dofine-dufei in https://github.com/langgenius/dify/pull/6620
- feat: added ActionButton component by @YIXIAO0 in https://github.com/langgenius/dify/pull/6631
- chore: optimize asynchronous deletion performance of app related data by @takatost in https://github.com/langgenius/dify/pull/6634
- chore: optimize asynchronous workflow deletion performance of app related data by @takatost in https://github.com/langgenius/dify/pull/6639
- doc: fix about model features by @zhangzhiqiangcs in https://github.com/langgenius/dify/pull/6619
- fix: value is not an array by @crazywoola in https://github.com/langgenius/dify/pull/6632
- Feat/user session id search by @ZhouhaoJiang in https://github.com/langgenius/dify/pull/6638
- remove rerank model must be required when retrieval_model is multiple by @JohnJyong in https://github.com/langgenius/dify/pull/6640
- feat(api/core/app/segments/variables.py): Support description in Variable. by @laipz8200 in https://github.com/langgenius/dify/pull/6636
- refactor(api/core/workflow/workflow_engine_manager.py): Remove by @laipz8200 in https://github.com/langgenius/dify/pull/6630
- Chores: add missing profile for middleware docker compose cmd and fix ssrf-proxy doc link by @vicoooo26 in https://github.com/langgenius/dify/pull/6372
- fix(api/core/workflow/nodes/iteration/iteration_node.py): Extend output in iteration if output is a array. by @laipz8200 in https://github.com/langgenius/dify/pull/6647
- fix reranking model field error by @JohnJyong in https://github.com/langgenius/dify/pull/6654
- Resolve variable type parameter error by @majian159 in https://github.com/langgenius/dify/pull/6646
- fix tencent_cos_storage image-preview error is not a byte by @yanghx-git in https://github.com/langgenius/dify/pull/6652
- [seanguo] feat: add llama 3.1 support in bedrock by @longzhihun in https://github.com/langgenius/dify/pull/6645
- chore: fix type annotations by @hjlarry in https://github.com/langgenius/dify/pull/6600
- Add support of tool-call for model provider "hunyuan" by @maybemaynot in https://github.com/langgenius/dify/pull/6656
- fix(api/fields/workflow_fields.py): Add check in environment variables by @laipz8200 in https://github.com/langgenius/dify/pull/6621
- feat: support max_retries in jina requests by @xielong in https://github.com/langgenius/dify/pull/6585
- Enhance database URI security and add URL encoding by @Seayon in https://github.com/langgenius/dify/pull/6668
- refactor(api/core/app/segments): implement
to_object
in ObjectVariable and ArrayVariable. by @laipz8200 in https://github.com/langgenius/dify/pull/6671 - fix bugs(when using Oracle23ai as Vector DB) by @tmuife in https://github.com/langgenius/dify/pull/6658
- fix: code block segmentation problem of markdown document by @greycodee in https://github.com/langgenius/dify/pull/6465
New Contributors
- @zhangzhiqiangcs made their first contribution in https://github.com/langgenius/dify/pull/6619
- @vicoooo26 made their first contribution in https://github.com/langgenius/dify/pull/6372
- @Seayon made their first contribution in https://github.com/langgenius/dify/pull/6668
Full Changelog: https://github.com/langgenius/dify/compare/0.6.15...0.6.16